dstack
@dstackai
An open-source alternative to Kubernetes and Slurm for AI container orchestration, supporting NVIDIA, AMD, TPU, and Intel across cloud and on-prem.
Introducing Fleets: a new open-source tool to manage cloud and on-prem clusters for AI. No more Kubernetes, KubeFlow, Slurm, etc. Fleets enable researchers and engineers to develop AI seamlessly, without friction with Ops teams. dstack.ai/docs/fleets/
In case you missed it: @nebiusai AI Cloud now supports dstack — a way friendlier alternative to Slurm and Kubernetes for ML teams. Dev environments, cluster training, inference — all covered. Docs 👉 docs.nebius.com/3p-integration…
Now that we have an API, we're officially on the @dstackai roadmap for direct support in provisioning a 1xMI300x vm. Thanks @andrey_cheptsov! github.com/dstackai/dstac…
How do @AMD GPUs perform in VMs vs bare-metal environments? We benchmarked containerized workloads on both — for training and inference using the MI300X. Read the results: dstack.ai/blog/benchmark…
If you want to be an AI developer and target those high paying gigs that are all the rage right now, you need access to enterprise compute. Saying that you trained a SOTA model on your home gaming GPU rig doesn’t sound nearly as impressive as saying that you setup @dstackai to…
Can VMs match bare-metal for GPU-intensive workloads? We put @AIatAMD VMs to the test to find out. Check out @dstackai's new benchmark series, supported by @HotAisle, now live: dstack.ai/blog/benchmark…
We asked @dstack and @andrey_cheptsov to do an unbiased investigation into whether or not our 1xMI300x virtual machines have any performance issues and this is what they discovered...
How do @AMD GPUs perform in VMs vs bare-metal environments? We benchmarked containerized workloads on both — for training and inference using the MI300X. Read the results: dstack.ai/blog/benchmark…
dstack 0.19.20 is out! The last release revamped log handling — this one fixes the UI issues that came with it. Log handling is now smoother, faster, and a lot more usable. 👇 github.com/dstackai/dstac…
New 0.19.19 release is out 🚀 Here are the changes it brings: * Update SSH fleets in-place. No more full redeploys to add/remove instances * Auto-cleanup unused volumes * Improved server performance — API responses up to 2x faster Full release notes: github.com/dstackai/dstac…
Amazing to see dstack featured by @nebiusai in their Q2 report! ML teams need better tools for @nvidia GPU orchestration — and open source plays a key role.
Here are all the Nebius AI Cloud updates we delivered in Q2: nebius.com/blog/posts/q2-… Welcome back to our quarterly roundup featuring all the improvements and updates released over the past three months. This overview highlights the key developments we’ve made to enhance your AI…
Announcing our new @AMD GPU benchmark! 🚀 1️⃣ Bare-metal vs. Containers: The "container tax" myth is busted. Performance is on par. 2️⃣ Partitioning: Performs well in isolation, struggles in cross-partition workloads. Full findings & charts inside: dstack.ai/blog/benchmark…
Still surprised how many ML teams avoid containers for training on large clusters. At @dstackai, we’ve started busting the myths—starting with @AMD GPUs. dstack.ai/blog/benchmark… Next up: the myth that container orchestration requires Kubernetes.
Dropped a new @dstackai benchmark on @AMD GPUs now on Hacker News. Help us get it in front of more devs! ☺️
🚀 Introducing Changelog! Thanks to your feedback, we’ve shipped some big improvements recently — Rolling deployments, Secrets, Files, @tenstorrent , Docker-in-Docker, @NVIDIA GPU support, and so much more. Read on more detail via dstack.ai/blog/changelog…
Say hello to AI container orchestration done right!
🚀 Introducing Changelog! Thanks to your feedback, we’ve shipped some big improvements recently — Rolling deployments, Secrets, Files, @tenstorrent , Docker-in-Docker, @NVIDIA GPU support, and so much more. Read on more detail via dstack.ai/blog/changelog…